January 30, 1997

Introduction

These comments are submitted on behalf of Citizens for a Better Environment (CBE). CBE is a regional environmental non-profit research and advocacy organization with over 8,000 members and thousands more supporters in Wisconsin.

In response to comments on the Draft Environmental Impact Statement (DEIS) from the Dane County Regional Planning Commission, U.S. EPA Region V, environmental organizations, and others, the Supplemental DEIS includes a much more substantial analysis of the indirect or secondary land use impacts of expanding USH 12 to four lanes. This analysis concludes that the highway expansion will result in a 4% population increase over the "no build" or "optimized two-lane" highway scenarios in the identified commutershed area for the 15-year period from 2005 to 2020. The 4% population increase is characterized as a conservative assumption.

For the reasons spelled out below, we believe the analysis is flawed and the projected 4% population increase for the period from 2005 to 2020 greatly underestimates the likely secondary land use impacts of expanding USH 12 to four lanes.

Two points should also be kept in mind concerning the analysis. First, the magnitude and seriousness of the impacts of induced development will continue to grow in future years beyond 2020--the study for the analysis. Second, the land use impacts go beyond increased population growth. The amount of land developed per capita will increase as a result of the auto-oriented sprawl development that the highway expansion will encourage. The landscape will be degraded by the attracted highway-oriented commercial "strip" development. The scattered, low-density residential development will create auto-dependent land use patterns that increase auto use, continuing the driving/sprawl cycle.

Summary of Critique of the Indirect Land Use Impacts Analysis

The secondary land use impacts analysis contains several flaws relating to the methodology and the assumptions used. The most serious flaw is the failure to control for the many non-transportation related factors that influence development in the regional growth analysis. The intense development pressures in the USH 12 commutershed area combined with the high level of latent demand due to current peak hour congestion levels create a situation unlike others in the project study area where highways have been expanded (e.g., U.S. 151 North to Columbus). This limits the usefulness of the regional growth analysis done comparing community growth rates with highway types serving them.

The most informative analysis would have been to look at growth rates in communities both prior to and after expansion of the highway serving the communities. For example, Verona (and the surrounding towns) would make a good case study. A comparison could be done looking at growth rates in Verona before and after expansion of U.S. 151 West and construction of the US 12 South Beltline. DeForest might be another good case study, looking at growth rates before and after expansion of U.S. 51.

Most of the studies that have been done on the impact of highway capacity expansion have focused on "generated traffic" or "induced travel." "Generated traffic" includes increased vehicle travel due to the attraction of trips from other routes, times, and modes, and increased tripmaking by existing residents. However, it also includes new or generated trips due to land use changes caused by improved auto access. In the long run (i.e., 10+ years), most generated traffic is due to land use and structural changes such as sprawl development, agglomeration of services, increased auto ownership, and reduced transportation choices. These land use and structural changes tend to have synergetic effects, such as diseconomies in transit service, resulting in increased automobile dependency. (See attached table describing the different types of generated traffic.)

The results of studies done on generated traffic indicate just how far off the Supplemental DEIS's 4% population increase projection likely is. Generated traffic is measured in terms of "elasticities," an economics concept used to describe the change in consumption of a good resulting from a change in the good's price. The concept can also be applied to non-financial costs such as time and comfort.

A recent University of California study calculated elasticities of vehicle travel with respect to road capacity to be 0.15-0.3, 0.3-0.4, and 0.4-0.6 for 4, 10, and 16 years respectively. This means that up to 60 out of each 100 additional roadway spaces are filled with generated traffic within 16 years. The researchers indicate higher rates of traffic generation are likely in urban areas due to increased latent demand.

A second University of California study used times series data from 1973 to 1990 for a set of fourteen metropolitan areas in California. The analysis focused on state highway VMT and on the relationship of VMT to the supply of state highways, as measured in lane miles. The study found that at the metropolitan level, a 1% increase in lane miles induces a 0.9% increase in VMT within five years. Another researcher, Kenneth Small, cites studies indicating that 50% to 80% of an increase in highway capacity is soon filled with generated traffic. A National Highway Institute transport investment guide recommends a default value of -0.5 for the elasticity of highway travel demand with respect to user costs (i.e., the percent change in demand due to a percent change in travel time or financial cost). A UK researcher, who reviewed numerous studies there, recommends use of an average value of -0.5 for the elasticity of travel demand with respect to travel time in the short term and up to -1.0 in the long term.

Based on the University of California study and input from other researchers, Litman has produced the following graph illustrating the range of percentages of likely traffic growth over time following highway capacity increases under different latent demand conditions:

The percentages of generated traffic range from 38% to 98% after 15 years. While some of this increased traffic is due to increasing tripmaking and longer trips, the vast majority is due to land use changes and increased auto dependency. It is difficult to translate these numbers into a percentage increase in population growth in an area affected by the highway. However, even if only one-third to one-half of the generated traffic is due to population growth, which is a conservative assumption, this implies a range of 12% to 33% population growth--much greater than the 4% projected in the Supplemental DEIS.

Specific Study Flaws Resulting in Underestimation of Land Use Impacts of Expanding USH 12 to Four Lanes

The following are specific flaws identified in the secondary land use impacts analyses done which tend to underestimate the potential impacts within the 15-year study period:

1. The baseline WDOA population projections should not have been used for the "no build" or "optimized 2-lane" alternatives.

The Supplemental DEIS uses the 1990-2015 population projections of the Wisconsin Department of Administration (WDOA) Demographic Services Center as the baseline for the "no build" or "optimized two-lane" scenarios. However, the WDOA forecasts are based on an extrapolation of "the momentum of historic population change in an area in the absence of physical or political restrictions," such as "deteriorating transportation service."

Under the assumption of an optimized two-lane highway, the level of service (LOS) on the highway--now at LOS D and E in some portions according to WisDOT--is projected by WisDOT to be at LOS F by the design year 2019. The increased peak period congestion and reduced travel times would clearly serve as a constraint on future development within the identified USH 12 commutershed area and future traffic growth on the highway. This is illustrated by the following graph which illustrates the self-limiting effect of congestion:

Source: Litman, Todd, Determining Generated Traffic External Costs, Victoria Transport Policy Institute, Victoria, BC Canada (1995).

The graph shows that as highway capacity is filled up and the level of service decreases, population and traffic growth declines. If capacity is added, traffic growth continues due to generated traffic.

The growth-constraining impact of increasing congestion may in fact already be occurring. Annual average daily traffic (ADT) on USH 12 decreased 5.5% from 1993 to 1994 after increases of 9.1% and 11.4% the previous two years. Annual ADT increased again in 1995, but only a small amount with total ADT still below 1993 levels.

Because of the growth-limiting effect of congestion, the assumed population forecasts for the "optimized two-lane" scenario should be decreased. WisDOT admits that not expanding USH 12 to four lanes "may serve as a constraint, redirecting development to other corridors within the Greater Madison Area, and "this could result in actual future population levels that are less than WDOA population projections." However, in its analysis of the land use impacts of expanding USH 12 to four lanes, WisDOT uses the WDOA forecasts as the baseline for the "optimized two-lane" scenario. As a result, the analysis underestimates the difference in population growth in the commutershed area between the two scenarios.

2. Inaccurate travel speeds were assumed for the travel time vs. commuter distribution analysis.

For the travel time vs. commuter distribution analysis, a travel speed of 50 mph (and 20 mph through Sauk City) was assumed for the two-lane highway scenario, while a travel speed of 60 mph was assumed for the expanded four-lane scenario. Use of a 50 mph peak hour speed for the two-lane scenario is inconsistent with the current LOS D/E and Year 2019 LOS F projected by WisDOT for the highway. If WisDOT's traffic projections are accurate, average peak hour travel speeds should drop below 50 mph in the next few years and well below 50 mph by 2019 under the optimized two-lane highway scenario.

At the same time, the assumed average travel speed of 60 mph for the expanded four-lane highway scenario is probably low, at least until 2010 or so when generated traffic is likely to increase congestion again. The posted speed limit is likely to be 60 or even 65 mph as it is now on U.S. Highway 151 North. People routinely travel 5-10 mph above the posted speed limit.

Because assumed average travel speed differential between the two-lane and four-lane scenarios is 5-10 mph smaller than it should be--perhaps even more for the later years--the travel time vs. commuter distribution analysis underpredicts the likely increase commuters under the four-lane scenario.

3. The projection of increased commuters due to USH 12 expansion derived from the travel time vs. commuter distribution analysis fails take into account the lack of data on commuters to suburban Madison communities.

The commuter distribution analysis is based on census information identifying the current percentage of workers commuting to job sites within the corporate limits of the City of Madison. The census information does not include workers commuting to Madison suburbs. The DEIS notes that this omission in the data weakens the analysis due to the presence of major employment centers in Middleton, Fitchburg, and Verona. The lack of data on commuters to the expanding employment center in Middleton is particularly problematic.

Despite this recognized weakness in the analysis, WisDOT fails to adjust up the population growth projections of the commutershed area communities for the four-lane highway scenario. The projections should be increased to factor in the likely impact of commuters locating outside Madison and working in suburban Madison communities, especially Middleton.

4. The Town of Berry and the Village of Prairie du Sac should not have been excluded from the travel time vs. commuter distribution analysis.

The Town of Berry and the Village of Prairie du Sac have a relatively higher percentage of residents commuting to Madison for work based on travel time. (See Fig. IV-3, p. IV-29 in the Supp. DEIS) Because of this, these two communities were excluded from the travel time vs. commuter distribution analysis "because the analysis would yield lower commuting percentages than what currently exist."

The high number of commuters living in these communities indicates the desirability of them. Reducing the commute time will simply make them even more attractive. We therefore question why they were excluded from the analysis.

5. The regional growth analysis fails to distinguish between the different types and uses of four-lane highways.

The land use impacts of expanding a highway depend on the type of highway facility, the time the facility was developed, and the type and number of interchanges and access points. The regional growth analysis done fails to account for the difference in highway types nor control for other non-transportation related variables. Three of the six four-lane highways examined as part of the analysis of highway type vs. community growth are interstate highways (IH 90/94 North, IH 90 East, and IH 94 East). The interstate was built for and routed to primarily serve intercity traffic, not Madison-bound traffic coming from growing population centers such as DeForest. Access to the interstate is also much more limited. This greatly reduces the meaningfulness of the analysis and tends to understate the impact from expanding USH 12.

6. An analysis looking at the growth rates in particular communities before and after highways were expanded from two to four lanes would provide a better indication of the impact of highway expansion on community growth.

The vast majority of growth within the commutershed area of Madison has been in the west/northwest area. This is reflected in Fig. IV-2 on p. IV-26 of the Suppl. DEIS showing the percentages of workers commuting Madison. Many factors, such as the beauty and natural resources of the area, have contributed to its high growth rate. Traffic congestion on the two-lane highways (USHs 12 and 14) connecting communities in this area to Madison, which might have served to constrain growth, has not been a major problem.

The historic higher growth rates of communities in this area compared to the lower historic growth rates of communities to the east of Madison (most of which are served by four-lane interstates) makes it difficult to draw good conclusions from regional growth analysis done. Because the communities with the highest historic growth rates are mostly served by two-lane highways, the analysis tends to understate the potential impact of expanding USH 12. Highway expansions serve as major enablers and accelerators of growth if other non-transportation factors affecting growth are present. In the case of the USH 12 corridor, these factors have contributed to high growth rates within the USH 12 commutershed area. Expansion of USH 12 is bound to accelerate the already high growth rates in this area.

It would have been much more informative to look at the growth rates of areas both before and after highways serving them were expanded from two to four lanes. DeForest for example, which was eliminated from the analysis, because Highway 51 was expanded within the study period (1980-1995) would have been a good test case.

Respectfully Submitted,

William Schaefer, Staff Attorney/Planner Citizens for a Better Environment 222 S. Hamilton Street, #4 Madison, WI 53703 PH (608) 251-2804 FAX (608) 255-4053

1. The only studies cited are 10-20 years old (see p. IV-19)

Good description of ways improved USH 12 will influence development:

1. Residential development related to commuting

DEIS tries to downplay, saying comparable housing is available substantially closer to Madison.

2. Commercial and industrial development related to improved access to state and national transportation system (via IH 90/94)

3. Commercial strip development along highway